The present invention relates generally to plasma immersion ion implantation. In particular, the present invention is directed to a process for minimizing ion implantation of undesired species and enhancing the stability and repeatability of desired dopant ions.
Ion implantation is a widely used technology preferred in the industry for doping semiconductor wafer work pieces. Generally, an ion beam implanter generates a beam of ions in a source chamber which are then directed with varying degrees of acceleration toward the work piece. The beam of ions is comprised of positively charged ions that impinge upon an exposed surface of a semiconductor wafer work piece thereby xe2x80x9cdopingxe2x80x9d or implanting the work piece surface with ions. The advantage to this process is that the user can selectively determine which ions are implanted by careful selection of an ionizable source gas.
A conventional ion implanter generally comprises three sections or subsystems: (i) an ion source for outputting an ion beam, ii) a beamline including a mass analysis magnet for mass resolving the ion beam, and (iii) a target chamber which contains the semiconductor wafer work piece to be implanted by the ion beam. The entire region between the ion source and the semiconductor wafer work piece is maintained at high vacuum to prevent formation of neutrals by collision of beam ions with residual gas molecules.
Ion sources in conventional ion implanters typically generate an ion beam by ionizing a source gas within a source chamber to produce a plasma at pressures of about 10xe2x88x923 to 10xe2x88x925 torr. Generally, the plasma is formed through collision of the source gas with electrons from an arc discharge, or cold cathode source. Typical examples of source gases include arsine (AsH3), vaporized antimony (Sb), phosphine (PH3), diborane (B2H6), boron trifluoride or trichloride (BF3 or BCl3), silicon tetrachloride (SiCl4), vaporized gallium (Ga), vaporized indium (In), ammonia (NH3), hydrogen (H2) and Nitrogen (N2). Positively charged ions are discharged from the ion source and directed along an evacuated beam path provided by the beam line toward the target chamber.
The ion beam produced in the ion source chamber is mass filtered such that substantially only the desired dopant species enter the target chamber. For example, ionized BF3 gas will dissociate into B++, B+ and BF2+ after collision with electrons from an arc discharge. The separation of ions by mass analysis is performed by adjusting a magnetic field about the beam line as the ions are discharged from the ion source. As the ions travel through the magnetic field, the strength and direction of the magnetic field of the analyzing magnet is set such that only ion species with a proper atomic weight, i.e., B+, are deflected at a proper radius of curvature to follow the desired beam line path to the target chamber. Thus, the undesired ions, i.e., B++ and BF2+, will be substantially filtered and prevented from reaching the target chamber. After filtration, the desired dopant ions are typically accelerated to the target chamber. Once in the target chamber, the desired dopant ions within the beam line strike the work piece and are implanted therein. Beam energies for conventional ion implantation generally range from 2-30 keV for low energy implanters to several MeV for high energy implanters.
Plasma immersion ion implantation (PIII) is an emerging technology that differs from the conventional ion implantation previously described in that the work piece to be selectively doped with ions is immersed within a plasma in the target chamber. Thus, unlike conventional ion implantation, the target chamber of plasma immersion ion implanters functions as both the processing chamber and the plasma source. Accordingly, PIII systems do not have mass filtering magnets for separating mass ions. Advantageously, this allows plasma immersion implantation of ions to occur at higher dose rates with much lower energies than that typically observed with conventional ion implantation. Typical energies employed in PIII systems range from about 0.1 KeV to about 10 KeV. Higher dose rates translate directly to increased throughput whereas lower energies minimize implant damage.
In a typical PIII system, a voltage differential is periodically established between a plasma and a platen holding the work piece to attract ions generated in the plasma toward the work piece. An ionizable gas is continuously fed into the chamber and subsequently ionized by an electromagnetic energy, either inductively or capacitively coupled, to form the plasma. A sufficient voltage differential will result in a pulsed ion implantation into the surface of the work piece. However, since the entire work piece is exposed to the plasma, all material exposed therein including any outgassed photoresist byproducts, may become ionized by the plasma and undesirably implanted into the wafer. Thus, there is no mass discrimination as is the case with conventional ion implantation wherein generation of ions occur in an ion source chamber spaced apart from the target chamber and are subsequently mass filtered prior to implantation into the work piece. Accordingly, it is important in PIII systems to keep the chamber free from materials that could come in contact with the plasma during operation, ionize and become implanted into the work piece.
The use of plasma immersion ion implantation presents significant problems since prior to ion implantation, an organic photoresist masking material is commonly used to protect areas of the work piece from ion implantation while allowing the charged ions to penetrate, or implant into the work piece in exposed regions to produce the desired doping characteristics in the work piece. Those skilled in the art will recognize and appreciate that as the charged ions impinge upon the work piece surfaces, the energetic ions react with the photoresist mask material to break bonds within the photoresist which subsequently results in outgassing of photoresist reaction by-products. As a result, outgassing of hydrogen, nitrogen, water vapor, hydrocarbons and other materials occur. The rate of outgassing is generally exponential. That is, outgassing is greatest during the initial bombardment of ions on the photoresist material and tapers off dramatically as implantation continues. In PIII systems, the outgassed material may then become exposed to the plasma and ionized resulting in, among others, implantation of undesired species that could affect the desired electrical properties for the work piece. Since most photoresists are organic, the photoresist outgas includes various hydrocarbons species. Of all species of the resist which outgas, carbon is of greatest concern during plasma immersion ion implantation since implanted carbon has been generally known to increase leakage currents and degrade generation lifetime in devices.
Additionally, in a manner similar to conventional ion implantation, the outgassed photoresist can cause pressure variations within the chamber which in turn can affect the quantity of impurity dopants implanted since measurement of implantation dose is essentially the same in PIII and conventional ion implanters. It is believed that the outgassing of photoresist can cause pressure gradients across the wafer, thereby influencing plasma density and doping rate across the wafer.
Moreover, ion implantation in general, is known to cause degradation of critical dimensions (CD) in the photoresist mask which can lead to a decrease in the yields of high end devices. For example, high energy and high current ion implantation can cause blistering, reticulation, burning and thermal flow of the resist, thereby effecting critical dimensions As such, subsequent etching or implant processes may result in etching or implanting into undesired areas of the underlying substrate.
Prior art approaches to the problems associated with photoresist outgassing during ion implantation generally focussed on making the resist more resistant to outgassing prior to implantation or etching. For example, hardening the photoresist prior to ion implantation is known to minimize outgassing and can be achieved by UV curing, e-beam curing, hard baking or a combination thereof However, these processes require additional time and different equipment sets, which in turn adds to the overall cost and throughput of the manufactured devices.
Accordingly, there is a need for improved methods for use with plasma ion immersion implantation to minimize the deleterious effects of photoresist outgassing. Additionally, it would be desirable if during plasma ion immersion implantation that the resist could be hardened in situ to minimize outgassing of the resist during ion implantation.
The present invention is directed to methods for use with a plasma immersion ion implantation (PIII) system wherein dopant ions implant into a surface of a substrate having an organic photoresist mask thereon. The present invention overcomes problems associated with photoresist outgassing occurring as a result of plasma immersion ion implantation. Moreover, the present invention addresses problems unique to plasma immersion ion implantation, i.e., ionization by the implant plasma of any material exposed therein such as photoresist outgas.
In one embodiment, the PIII process includes placing the substrate having the organic photoresist mask thereon into a process chamber and pretreating the substrate prior to the steps of generating the implant plasma so as to substantially prevent implantation of undesired ions resulting from a reaction of the implant ions with the photoresist. Preferably, the substrate is a silicon semiconductor wafer. The pretreatment steps includes generating a plasma from an ionizable source to produce electrically inactive ions or electrons. The electrically inactive ions produced in the plasma react with the organic photoresist mask to cause outgassing. The outgas material is removed by evacuating. Preferably, the evacuation is performed as the outgassing material evolves from the wafer. Advantageously, this prevents subsequent implantation of any undesired species that may occur during the implant process steps. The source for the ionizable gas in the pretreatment process is a selected one of a noble gas and vaporizable material. Preferably, the noble gas is selected from the group consisting of helium, neon and argon and flows into the chamber at a rate greater than 50 sccm. The high flow rate and pumping speed cause the outgassed photoresist to have a small partial pressure and therefore contribute very little to the implant. In the case of argon, the electrically inactive species implanted into the substrate are 40Ar+. Preferably, the pretreatment steps are completed in less than about 10 seconds. The optical emission signal of carbon species produced during the outgassing may be monitored for determining when the carbon emission signal is no longer detectable. The steps of implanting electrically active ions doccurs once the outgassing is substantially complete or when the emission signal of carbon ions is no longer detectable.
An implant plasma of ions from an ionizable source is generated in a chamber wherein the pretreated substrate having the patterned photoresist thereon is implanted with electrically active ions when implanted. Voltage pulses are applied between the chamber walls and the substrate independently of the forming of the plasma for selectively attracting and implanting the electrically active ions into a surface of the substrate. Preferably, the source for the implant plasma is an ionizable gas and flows into the chamber at a rate of about 0.5 sccm to about 10 sccm. Preferably, the source for the ionizable gas is selected from the group consisting of arsine, vaporized antimony, phosphine, diborane, boron trifluoride or trichloride, silicon tetrachloride, vaporized gallium, vaporized indium, ammonia, hydrogen and nitrogen.
In another embodiment the process for use with PIII systems include the following sequence of steps:
a) placing the substrate having an organic photoresist pattern thereon into a chamber;
b) flowing a first ionizable gas into the chamber;
c) ionizing the first gas in the chamber to produce electrically inactive ions when implanted in the substrate;
d) reacting the electrically inactive ions with the photoresist to produce an outgas material and continuously evacuating the outgassed material from the chamber so as to substantially prevent ionizing the outgassed material;
e) turning off the flow of the first gas;
f) flowing a second gas into the chamber;
g) ionizing the second gas in the chamber to produce electrically active ions;
h) selectively implanting the electrically active ions into the substrate surface;
i) turning off the second gas flow; and
j) unloading the substrate from the chamber.
In another embodiment, a process for curing the photoresist prior to ion implantation is disclosed. The process includes placing a substrate having an organic photoresist patterned thereon in an evacuated chamber. A gas is ionized to produce positive ions and electrons. A positive voltage pulse on the platen attracts and accelerates electrons to the substrate. This electron processing cures the photoresist making it more resistant to damage during subsequent processing.
Further features and advantages of the present invention will become apparent from the following detailed description and accompanying drawings which form a part of the specification.